|dc.description.abstract||The use of Artificial Intelligence techniques in the development of educational software
brought the hope of developing systems that would become personalised to each
student and thus be of more benefit to him or her. But despite their added complexity,
these Intelligent systems (ITSs, ILEs, ICALLs, etc.) do not always succeed in
engaging the student. While a lot of effort has been spent investigating how to accommodate
an instructional interaction to the student's knowledge, almost no work has
been done in trying to accommodate the instruction to the student's motivational state.
This is surprising, given the immense impact that a student's motivation has in his or
The little previous work dealing explicitly with motivation in tutoring systems has
focused mainly on the strategies that an Intelligent Tutoring System (ITS) could use
to motivate the student. In this dissertation we focus on the prior (but we believe, fundamental),
task of detecting the student's motivational state, on which the mentioned
strategies could be used.
We argue that the available theories of motivation in education are not specific
enough and are of limited usefulness in order to implement a motivation detection
component in an ITS. Thus, we argue for the need of empirical studies that can help us
elicit formalised motivation diagnosis knowledge. To this effect, we discuss a number
of empirical studies we performed in order to inform the design of an ITS simulation
that detects the motivational state of a student.
The main aspects of the motivation diagnosis architecture presented in this dissertation
are a motivation self-report component and a motivation diagnosis component
based on human teachers' motivation diagnosis knowledge, elicited via one of the mentioned
empirical studies. This architecture was implemented as an ITS simulation in
order to help us evaluate these motivation diagnosis techniques.
The evaluation showed that, although not perfect, the motivation diagnosis techniques
introduced in this dissertation seem to offer a reasonable level of accuracy in
detecting a student's motivational state, and although the approach presented is not the
only possible one and many aspects of this work can still be improved, we believe that
it offers a promising step towards tutoring systems that care!||en